There are conflicting opinions as to whether the use of "gateway" drugs, such as nicotine, ethanol and delta-9-tetrachydrocannabinol lead directly to the use of "harder" drugs, such as cocaine, amphetamines and opioids. Furthermore, there are few data available on the question of how the introduction of a second drug of abuse quantitatively affects the pattern of intake of a drug for which self administration is already well established. The present experiments are designed to answer such questions using a rat model of drug self-administration. Hopefully, the findings of the present experiments will provide important insights as to the emergence of polydrug abuse patterns in humans. In initial experiments, rats will be trained to consume 5% (w/v) ethanol with water freely available by using periods of limited access to a schedule of food-pellet presentation. Preliminary data show that the rats take almost all of their fluid as ethanol using this procedure. One food, water and ethanol intake stabilize, the effects of acute and chronic administration of nicotine, delta-9- tetrahydrocannabinol and cocaine will be determined. Subsequently, chronic administration of each of these drugs will be discontinued to determine the effects of their withdrawal on ethanol self administration. In another experiment, rats will be trained to self administer intravenously, cocaine, nicotine and if possible, delta-9-tetrahydrocannabinol to determine to what extent the "voluntary" use of a second drug influences ethanol self administration. subsequently, the opportunity to self administer these drugs will be discontinued to determine the effects of this type of withdrawal on ethanol self administration. In other experiments, comparisons will be made between rats with and without a history of ethanol administration in the acquisition of patterns of nicotine, cocaine and possibly delta-9- tetrahydrocannabinol self administration. These experiments should provide considerable useful information about drug interactions during the development of polydrug abuse patterns.